This second revision of an R21 application is focused on a novel approach to the study of the neurobiology of marijuana use in subjects with and without bipolar disorder. The proposed investigation is designed to examine the extent to which structural changes in the frontal cortex and the functional response of the anterior cingulate are associated with poor inhibitory function and mood fluctuation in subjects with chronic heavy cannabis use and cannabis users with bipolar disorder. In addition, we will examine whether these frontal changes are predictive of behavior. Several studies have shown that bipolar patients have extremely high rates of comorbid substance use, and that cannabis is often used to alleviate severe mood symptoms. In recent work from our laboratory, we have found alterations in diffusion tensor imaging (DTI) measures of frontal white matter as well as BOLD changes in anterior cingulate (AC) and dorsolateral prefrontal activity (DLPFC) during the completion of an inhibitory challenge paradigm in chronic heavy marijuana (MJ) smokers. We have also reported alterations in functional activity in this region in patients with bipolar disorder. These findings complement previous reports and suggest that this region may play a crucial role in the modulation of inhibitory functions in both disorders. To date, no investigation has examined these neurobiologic correlates with regard to their potential predictive value regarding mood changes and number of episodes of smoking. We will apply DTI and blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) techniques to measure differences in white matter and regional activation in focal brain areas during cognitive and affective challenge paradigms in 20 chronic, heavy MJ smokers, 20 chronic heavy MJ smokers with Bipolar I disorder, 20 Bipolar patients who do not smoke MJ, and 20 healthy non-smoking control subjects. We also propose to examine "real time" mood changes and cannabis use with handheld electronic devices designed for use within subjects' own environments. This data will be used to test the hypotheses that: 1) BOLD signal changes will be different in MJ smokers with and without bipolar disorder relative to control subjects; 2) DTI measure of white matter will be altered in MJ smokers with and without bipolar disorder; and 3) fMRI BOLD and DTI measures will predict changes in mood, measures of impulsivity and frequency of smoking in a four week period following the scanning session. To our knowledge, no studies have utilized neuroimaging measures as predictors of behavior. Data on behavior and ratings of mood will be collected in "real time" within subjects' own environment, using a handheld PDA device. This investigation is innovative, in that it includes data acquired from fMRI and DTI techniques as predictors of real time behavior in substance abusers. We believe these findings may help identify potential risk factors for substance use disorders, as well as guide the development of new treatment interventions. The proposed investigation is designed to examine the extent to which structural changes in the frontal cortex and the functional response of the anterior cingulate cortex are associated with poor inhibitory function and mood fluctuation in subjects with chronic heavy cannabis use with and without bipolar disorder. In addition, we will examine whether frontal changes are predictive of mood alterations and frequency of smoking using "real time" assessments of mood and behavior. The unique, previously untapped combination of structural, functional, and diffusion tensor imaging methods applied to these populations allows for rigorous examination of specific cortical regions involved in cannabis dependence, allowing for the identification of potential risk factors for substance abuse and may provide important information regarding the development of new treatment interventions. [unreadable] [unreadable] [unreadable]